System and Method for Modeling In-Train Forces with Multiple Locomotives Interspersed Evenly Along Train

1. A train control system comprising: a plurality of locomotives spaced along a length of a train consist, each locomotive of the plurality of locomotives comprising: one or more independent virtual in-train forces modeling engines configured to simulate in-train forces and train operational characteristics; and an analytics engine and a calibration engine configured to assimilate, analyze, and calibrate real-time information from: one or more of the plurality of locomotives, sensors acquiring real-time data on forces and displacements at draft gears and couplers interconnecting the plurality of locomotives with non-powered rail cars in between each of the plurality of locomotives, determinations made by the one or more independent virtual in-train forces modeling engines, and an energy management system associated with each of the plurality of locomotives of the train and configured to determine one or more of throttle requests, dynamic braking requests, and pneumatic braking requests for the locomotive of the plurality of locomotives based at least in part on the one or more independent virtual in-train forces modeling engines, the analytics engine, and the calibration engine, wherein: the energy management system is configured to operate independently of control signals from a lead locomotive or central command, each of the independent virtual in-train forces modeling engines being configured to model a segment of the train including the locomotive of the plurality of locomotives and the non-powered rail cars connected to the locomotive using a machine learned system trained using historical operational and structural data comprising characteristics of the segment of the train, changes in speed of the locomotive, and in-train forces of the segment of the train, each of the independent virtual in-train forces modeling engines being configured to simulate in-train forces and train operational characteristics using: physics-based equations, kinematic or dynamic modeling of behavior of the train or components of the train during operation when at least one of the plurality of locomotives is changing speed, and inputs derived from stored historical contextual data comprising characteristics of the components of the train and a track over which the train is traveling.

2. The train control system of claim 1, wherein the sensors acquiring real-time data include one or more of LIDAR sensors, RADAR sensors, accelerometers, gyroscopic sensors, optical recognition sensors, and physical strain gauges configured to measure displacements or forces at the draft gears and couplers interconnecting the plurality of locomotives, and each of the analytics engines is configured to fuse the data acquired by each of the sensors.

3. The train control system of claim 1, wherein the machine learning used by each of the independent virtual in-train forces modeling engines includes implementation of at least one of a neural network, a support vector machine, a Markov decision process engine, a decision tree based algorithm, or a Bayesian based estimator.

4. The train control system of claim 3, wherein the machine learning includes implementation of a neural network, including training the neural network by providing the inputs derived from stored historical contextual data as an input to a first layer of the neural network, wherein an output generated by a learning function during implementation of the machine learning includes a plurality of first outputs from the neural network generated based on the inputs, and at least one learning parameter includes a characteristic of the neural network which is modified to reduce a difference between the plurality of first outputs and the historical operational and structural data.

5. The train control system of claim 1, wherein each of the virtual in-train forces modeling engines is configured to simulate the in-train forces and train operational characteristics during a period of time when the train includes at least one locomotive with an associated energy management system that is changing the speed of the at least one locomotive.

6. The train control system of claim 1, wherein the real-time and historical operational and structural data acquired by each of the independent virtual in-train forces modeling engines for use as training data includes one or more of structural stresses on one or more couplers or draft gears interconnecting locomotives or locomotives and non-powered rail cars of the train.

7. The train control system of claim 6, wherein the energy management system is configured to adjust one or more of throttle requests, dynamic braking requests, and pneumatic braking requests using a microprocessor-based locomotive control system, a cab electronics system, and an electronic pneumatic brake system mounted within a cab of each of the one or more locomotives.

8. The train control system of claim 7, wherein the energy management system is configured to adjust one or more of throttle requests, dynamic braking requests, and pneumatic braking requests for the train while transitioning and ramping up from heavy dynamic braking at a bottom of a hill to full throttle back up an adjacent hill in a direction of travel of the train along the train track, and while increasing a ramp rate and maintaining the structural stresses on one or more of the couplers or draft gears.

9. The train control system of claim 1, wherein each of the independent virtual in-train forces modeling engines is configurable by a user in order to adjust relative weights that are assigned to each of different types of the acquired real-time and historical operational and structural data used as training data for each modeled segment of the train, and wherein predetermined acceptable range of values for in-train forces and train operational characteristics are configurable by the user.

10. A method of using independent virtual in-train forces modeling engines disposed onboard each of a plurality of locomotives that are spaced at approximately regular intervals along an entire length of a train in order to model segments of the train, each segment including a locomotive of the plurality of locomotives and associated non-powered rail cars connected to the locomotive, and to coordinate accelerations and decelerations of each of the plurality of locomotives determined independently onboard each of the plurality of locomotives, the method comprising: assimilating, analyzing, and calibrating real-time information from: one or more of the plurality of locomotives, sensors acquiring real-time data on forces and displacements at draft gears and couplers interconnecting the plurality of locomotives with non-powered rail cars in between each of the plurality of locomotives, and determinations made by the independent virtual in-train forces modeling engines, operating each of the plurality of locomotives of the train collectively to minimize in-train forces independently of command and control signals from a lead locomotive and central command, modeling a segment of the train including one of the plurality of locomotives and the non-powered rail cars connected to the locomotive with each of the independent virtual in-train forces modeling engines using acquired real-time and historical operational and structural data comprising characteristics of the locomotive and using machine learning performed by a machine learning engine for controlling changes in speed of the locomotive of the plurality of locomotives on which it is disposed independently of command and control signals from the lead locomotive and central command, simulating in-train forces and train operational characteristics with each of the independent virtual in-train forces modeling engines using: physics-based equations, kinematic or dynamic modeling of behavior of the train or components of the train during operation when at least one of the plurality of locomotives is changing speed, and inputs derived from stored historical contextual data comprising characteristics of the components of the train and a track over which the train is traveling, and using an energy management system associated with each of the plurality of locomotives of the train, adjusting one or more of throttle requests, dynamic braking requests, and pneumatic braking requests for each of the plurality of locomotives based at least in part on real-time information from one or more of the plurality of locomotives, from the draft gears and couplers, and from determinations made by the independent virtual in-train forces modeling engine disposed onboard each of the plurality of locomotives.

11. The method of claim 10, wherein the sensors acquiring real-time data include one or more of LIDAR sensors, RADAR sensors, accelerometers, gyroscopic sensors, optical recognition sensors, and physical strain gauges configured to measure displacements or forces at the draft gears and couplers, and the method further comprises fusing the data acquired by each of the sensors using an analytics engine.

12. The method of claim 10, wherein the machine learning includes implementation of a neural network, including training the neural network by providing the inputs derived from stored historical contextual data as an input to a first layer of the neural network, wherein an output generated by a learning function during implementation of the machine learning includes a plurality of first outputs from the neural network generated based on the inputs, and at least one learning parameter includes a characteristic of the neural network which is modified to reduce a difference between the plurality of first outputs and training data.

13. The method of claim 10, wherein each of the virtual in-train forces modeling engines is configured to simulate the in-train forces and train operational characteristics during a period of time when the train includes at least one locomotive with an associated energy management system that is changing the speed of the at least one locomotive.

14. The method of claim 13, wherein the real-time and historical operational and structural data acquired by each of the independent virtual in-train forces modeling engines for use as training data includes one or more of structural stresses on one or more couplers or draft gears interconnecting locomotives or locomotives and non-powered rail cars of the train.

15. The method of claim 10, wherein the energy management system is configured to adjust one or more of throttle requests, dynamic braking requests, and pneumatic braking requests for the one or more associated locomotives of the train using a microprocessor-based locomotive control system, a cab electronics system, and an electronic pneumatic brake system mounted within a cab of each of the one or more locomotives.

16. The method of claim 15, wherein the energy management system is configured to adjust one or more of throttle requests, dynamic braking requests, and pneumatic braking requests for the one or more associated locomotives of the train while transitioning and ramping up from heavy dynamic braking at a bottom of a hill to full throttle back up an adjacent hill in a direction of travel of the train along a train track, and while increasing a ramp rate and maintaining the structural stresses on one or more of the couplers or draft gears and vibrations of engine components within predetermined acceptable ranges of values.

17. The method of claim 10, wherein each of the independent virtual in-train forces modeling engines is configurable by a user in order to adjust relative weights that are assigned to each of different types of the acquired real-time and historical operational and structural data used as training data for each modeled segment of the train, and wherein predetermined acceptable range of values for in-train forces and train operational characteristics are configurable by the user.

18. The method of claim 10, wherein the machine learning used by each of the independent virtual in-train forces modeling engines includes implementation of at least one of a neural network, a support vector machine, a Markov decision process engine, a decision tree based algorithm, or a Bayesian based estimator.

19. A locomotive control system, comprising: a learning system configured to: receive real-time and historical operational and structural data from one or more sensors of a segment of a train consist including a plurality of locomotives, the segment of the train consist including a locomotive and one or more non-powered rail cars; train one or more independent virtual in-train forces modeling engines configured to model the segment of the train consist using the historical operational and structural data comprising characteristics of the segment of the train, changes in speed of the locomotive, and in-train forces of the segment of the train, wherein the virtual in-train forces modeling engines are disposed onboard the locomotive; simulate, using each of the independent virtual in-train forces modeling engine engines, in-train forces and train operational characteristics using: physics-based equations, kinematic or dynamic modeling of behavior of the train or components of the train during operation when at least one of the plurality of locomotives is changing speed, and inputs derived from stored historical contextual data comprising characteristics of the components of the train and a track over which the train is traveling; store one or more independent virtual in-train forces models simulated by the virtual in-train forces modeling engine in a model database, wherein each of the one or more independent virtual in-train forces models includes a mapping between different combinations of the stored historical contextual data and corresponding simulated in-train forces and train operational characteristics that occur when the locomotive is changing speed; calculate, using a machine learning engine, relative weights to assign to each of different types of the stored historical contextual data used by each of the independent virtual in-train forces modeling engines and assign the relative weights to the stored historical contextual data; train the learning system with the machine learning engine using the weighted stored historical contextual data and training data to determine a probability of each of the independent virtual in-train forces models providing an accurate representation of actual in-train forces and train operational characteristics that occur when one or more of the plurality of locomotives changes speed, and using a learning function including at least one learning parameter, wherein training the learning system includes: providing the weighted stored historical contextual data as an input to the learning function, the learning function being configured to use the at least one learning parameter to generate an output based on the input; causing the learning function to generate the output based on the input; comparing the output to the training data, wherein the training data includes data produced by sensors having captured actual information on in-train forces and train operational characteristics during the changes of speed of the one or more locomotives of the plurality of locomotives; comparing the determined probabilities of each of the virtual in-train forces models to a predetermined threshold probability level; and initiating adjustments to one or more of the relative weights assigned to each of the different types of the stored historical contextual data of each of the independent virtual in-train forces models to improve the determined probabilities of each of the independent virtual in-train forces models based on actual information on in-train forces and train operational characteristics acquired from a plurality of different trains operating under different conditions; and adjust one or more of throttle requests, dynamic braking requests, and pneumatic braking requests for each of the one or more of the plurality of locomotives of the train that is changing speed using an energy management system associated with the one or more locomotives of the train based at least in part on one of the virtual in-train forces models with the highest probability of providing an accurate representation of actual in-train forces and train operational characteristics and with one or more of the simulated in-train forces and train operational characteristics falling within a predetermined acceptable range of values.

20. The locomotive control system of claim 19, wherein the training data includes configuration and operational data associated with the inputs derived from stored historical contextual data, the training data being generated by one or more systems or components of the train while the train is being operated by a subset of train operators selected based at least in part on experience, wherein the output generated by the learning function represents an objective that the machine learning engine is configured to cause the learning system to match by modifying the at least one learning parameter until the difference between the output and the training data is less than a predetermined threshold difference.